Neuroscientists are faced with a torrent of experimental data about the structure, function, and development of the brain in health and disease. To aid in coping with this flood, the present project aims to provide the neuroscience community with (i) a well-integrated set of software tools for visualizing, analyzing, accessing, and communicating information about the cerebral and cerebellar cortex and (ii) a set of surface-based atlases that provide a compendium of information about human, monkey and rodent cortex. One objective is to implement a unified software application (Caret) that will carry out fully automated segmentation (to capture the shape of the cortex in individual brains), plus multiple stages of surface-based analysis. These analyses will include generating cortical flat maps and spherical maps, identifying cortical sulci, mapping cortical thickness, and registering individuals to the atlas map. A second objective is to enhance the Surface Management System (SuMS) database, by incorporating powerful search capabilities, and by improved methods for visualizing search results, both online and offline. SuMS will be a distributed database network that allows local file storage with multiple security levels as well as access to the central SuMS repository. A third objective is to improve the methods for surface-based registration of one cerebral hemisphere to another, in order to better compensate for individual variability within a species and to provide improved methods for making comparisons across species. A fourth objective is to expand the mapping of experimental data from a variety of sources onto surface-based atlases of human, macaque, rat, and mouse cerebral and cerebellar cortex. A major focus will be on the development of probabilistic maps of visual areas in monkey and human cortex. Attainment of these overall objectives will allow neuroscientists everywhere to access many types of experimental information about cerebral and cerebellar cortex with much greater ease and flexibility than is currently possible.